自抗扰控制器的优化设计及应用研究

发布时间：2018-05-28 03:02

本文选题：自抗扰控制 + 免疫遗传算法　；参考：《华北电力大学》2014年硕士论文

【摘要】：随着我国电力工业的快速发展,热工过程中被控对象维数越来越高、结构越来越复杂,致使系统控制难度越来越大。为了提高系统性能,各类先进控制算法应运而生,如鲁棒控制、预测控制、自适应控制及分散控制等,虽然这些算法都能改善系统的性能,但结构比较复杂,计算量较大。自抗扰控制(ADRC)结构简单、易于设计、抗干扰能力强,自上世纪八十年代被提出以来,已经取得了突破性的进展,迄今为止仍然是控制理论研究的中心问题之一。本文主要研究自抗扰控制器的优化设计及应用研究,努力做到结构优化与参数优化并重,改善系统的性能。首先,在非线性自抗扰控制的基础上,针对其参数多不易调节的缺点,本文将免疫理论中的非特异性免疫、特异性免疫与遗传算法结合,形成了免疫遗传算法。该算法不仅继承了遗传算法与免疫算法的优点,还提高了算法的稳定性、收敛精度和收敛速度。并通过变速恒频风力发电系统自抗扰控制器参数优化实例证明该算法的有效性,解决控制器参数优化问题,提高风能利用效率。其次,针对实际生产过程中多变量系统普遍存在耦合的特点,本文首先采用不变性原理设计多变量系统的动态解耦补偿器；为了简化控制模型,进一步对补偿器模型进行简化；其次在动态解耦的基础上,引入自抗扰控制(ADRC),形成自抗扰解耦控制方案,利用ADRC的解耦功能,将变量间的耦合看作一种扰动进行实时估计和补偿,实现进一步解耦；然后,以循环流化床系统为例,设计该系统解耦前、解耦后的Matlab仿真实验,并与PID控制的仿真结果进行对比,验证该方案的有效性。最后,鉴于工业控制系统中普遍存在着迟延现象,本文充分利用Smith结构对迟延的补偿优势以及ADRC抗扰性强的优点,将Smith结构和ADRC相结合,形成既能补偿迟延又能抑制扰动的史密斯-自抗扰控制(Smith-ADRC)方案。该方案结构简单,设计简便,且具有很强的鲁棒性和适应性。然后,以单变量的主汽温系统、多变量的球磨机系统为例设计仿真实验,并与Smith-PID、ADRC的优化结果进行对比,验证该方案的有效性。
[Abstract]:With the rapid development of electric power industry in our country, the dimension of the controlled object is getting higher and the structure is more and more complex in the process of thermal engineering, which makes the control of the system more and more difficult. In order to improve the performance of the system, various advanced control algorithms emerge as the times require, such as robust control, predictive control, adaptive control and decentralized control. Although these algorithms can improve the performance of the system, the structure is complex and the calculation is large. ADRC (active disturbance rejection Control) has a simple structure, easy design and strong anti-interference ability. Since it was proposed in the 1980s, it has made a breakthrough and is still one of the central problems in the control theory research up to now. In this paper, the optimal design and application of the ADRC are studied, and the structural optimization and parameter optimization are emphasized to improve the performance of the system. Firstly, on the basis of nonlinear ADRC, the immune genetic algorithm (IGA) is formed by combining the non-specific immunity, specific immunity and genetic algorithm in the immune theory, aiming at the disadvantage that the parameters are not easy to adjust. The algorithm not only inherits the advantages of genetic algorithm and immune algorithm, but also improves the stability, convergence accuracy and convergence speed of the algorithm. An example of parameter optimization of active disturbance rejection controller for variable speed constant frequency wind power generation system is given to prove the effectiveness of the algorithm, to solve the problem of controller parameter optimization and to improve the efficiency of wind energy utilization. Secondly, in order to simplify the control model, the dynamic decoupling compensator of multivariable system is designed based on invariance principle in order to simplify the control model. Secondly, on the basis of dynamic decoupling, the ADRC controller is introduced to form the ADRC control scheme. By using the decoupling function of ADRC, the coupling between variables is estimated and compensated as a disturbance in real time to realize further decoupling. Taking the circulating fluidized bed system as an example, the Matlab simulation experiment before and after decoupling is designed, and the simulation results of PID control are compared to verify the effectiveness of the scheme. Finally, in view of the widespread phenomenon of delay in industrial control system, this paper makes full use of the advantage of Smith structure to compensate delay and the advantage of strong ADRC immunity, and combines Smith structure with ADRC structure. A Smith-ADRC scheme, which can compensate delay and suppress disturbance, is formed. The structure of the scheme is simple, the design is simple, and it has strong robustness and adaptability. Then, the simulation experiment is designed with the single-variable main steam temperature system and the multi-variable ball mill system as an example, and compared with the optimized results of Smith-PIDN ADRC, the effectiveness of the scheme is verified.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM571

【参考文献】